Analysis
Mueon Corporation, a Portland, Oregon semiconductor systems company, has raised $65 million in a Series A led by Maverick Silicon to scale what it calls a modular, exponential computing platform for AI and hyperscale data centers, the company announced Thursday. Intel Capital, the Geodesic Alliances Fund, Oregon Venture Fund, M-Ventures, Wellington Capital, IQT and Defined also participated.
From stealth to $80 million raised
Mueon emerged from stealth with a $15.5 million seed round in 2025; the new Series A brings its total funding to more than $80 million in roughly a year and a half. As part of the round, Manish Muthal joins Mueon's board alongside Intel Capital managing director Srini Ananth -- a signal that Intel Capital, as a strategic rather than purely financial investor, wants direct influence over how Mueon's architecture develops.
“That's the backdrop pulling venture dollars toward companies like Mueon that target the physical-assembly problem rather than the chip itself.”
The problem Mueon says it's solving
Mueon's pitch is structural rather than incremental: instead of building faster individual chips, the company is redesigning how compute, memory and interconnect are physically assembled into what it calls "Cubelets" -- modular units meant to be reconfigured as AI workloads scale, rather than requiring a full data-center redesign. That puts Mueon in competition with a crowded field of AI-infrastructure startups chasing the same bottlenecks, including Lightmatter, Celestial AI and photonics player Salience Labs, which raised a $125 million Series B this same week for a related but distinct approach to AI data-center networking.
Why the money is flowing here
Data-center power delivery and physical layout, not raw chip performance, have become the real limiting factor on AI buildouts industry-wide, with chipmakers and hyperscalers increasingly flagging packaging and power as tighter bottlenecks than wafer supply heading into 2027. That's the backdrop pulling venture dollars toward companies like Mueon that target the physical-assembly problem rather than the chip itself.
The risk for Mueon: a $65 million Series A is a modest sum next to the billions hyperscalers and chip giants are pouring into their own in-house interconnect and packaging research, and a modular hardware thesis has to prove out in actual deployed data centers, not pitch decks, before it can compete with vertically integrated rivals building the same capability internally.